Method for searching contour noise points of three-dimensional point cloud data

Abstract

The invention provides a method for searching contour noise points of a three-dimensional point cloud data, and the method is used for searching the contour surface noise points of the three-dimensional point cloud data based on an optical surface scanning three-dimensional measurement system. The method comprises the following steps of: searching and cancelling remote noise points on the contour surface of the three-dimensional point cloud data; and computing the normal vectors of all points in the three-dimensional point cloud data, searching the contour noise points of the three-dimensional point cloud data, etc. According to the method, a good man-machine interaction function can be realized, and the contour noise points of the three-dimensional point cloud data can be efficiently and exactly searched, so that an effective method for improving the quality of the point cloud data can be provided. When the method is specifically realized, the searched noise points are firstly marked to be the color which is different from the other data points, to be distinguished, and are stored in a special dynamic array. Finally, if the noise points are confirmed to be capable of being cancelled, the noise points which are marked and stored in the dynamic array are removed. Therefore, the fairness of the contour surface of the point cloud data can be effectively improved, and the stereoscopic vision effect can be improved.

Description

technical field [0001] The invention relates to the field of stereo vision point cloud data processing methods, in particular to methods for searching and removing contour noise points of massive three-dimensional point cloud data. Background technique [0002] A point cloud is a collection of massive points expressing the spatial distribution of objects and surface properties of objects in the same spatial reference system. The point cloud obtained according to the principle of close-range photogrammetry, including three-dimensional coordinates (XYZ). The properties of point cloud include: spatial resolution, point accuracy, surface normal vector, etc. After obtaining the spatial coordinates of each sampling point on the surface of the object, a collection of points is obtained, which is called "Point Cloud". [0003] Point cloud data (point colud) refers to the form of data obtained by an optical 3D scanner. The scan data is recorded in the form of points, each point co...

AI Technical Summary

Problems solved by technology

[0004] Due to the influence and limitation of the 3D optical measurement equipment by the measurement environment, the shape of the measured entity, etc., during the measurement process, it is inevitable that there will be more noise points or noise point slices in the measurement data. These noise points not only affect the measurement The accuracy of the result, and it will also cause a lot of trouble for the subsequent processing of the point cloud

Therefore, in order to eliminate noise points, processing and control can be carried out from all links before point cloud generation, but it is difficult to take into account all measurement conditions through such processing and improvement

Images